Large Language Models (LLMs) sometimes suffer from producing hallucinations, especially LLMs may generate untruthful responses despite knowing the correct knowledge. Activating the truthfulness within LLM is the key to fully unlocking LLM's knowledge potential. In this paper, we propose TruthX, an inference-time intervention method to activate the truthfulness of LLM by identifying and editing the features within LLM's internal representations that govern the truthfulness. TruthX employs an auto-encoder to map LLM's representations into semantic and truthful latent spaces respectively, and applies contrastive learning to identify a truthful editing direction within the truthful space. During inference, by editing LLM's internal representations in truthful space, TruthX effectively enhances the truthfulness of LLM. Experiments show that TruthX improves the truthfulness of 13 advanced LLMs by an average of 20% on TruthfulQA benchmark. Further analyses suggest that TruthX can control LLM to produce truthful or hallucinatory responses via editing only one vector in LLM's internal representations.

While current large language models (LLMs) demonstrate some capabilities in knowledge-intensive tasks, they are limited by relying on their parameters as an implicit storage mechanism. As a result, they struggle with infrequent knowledge and temporal degradation. In addition, the uninterpretable nature of parametric memorization makes it challenging to understand and prevent hallucination. Parametric memory pools and model editing are only partial solutions. Retrieval Augmented Generation (RAG) $\unicode{x2013}$ though non-parametric $\unicode{x2013}$ has its own limitations: it lacks structure, complicates interpretability and makes it hard to effectively manage stored knowledge. In this paper, we introduce MemLLM, a novel method of enhancing LLMs by integrating a structured and explicit read-and-write memory module. MemLLM tackles the aforementioned challenges by enabling dynamic interaction with the memory and improving the LLM's capabilities in using stored knowledge. Our experiments indicate that MemLLM enhances the LLM's performance and interpretability, in language modeling in general and knowledge-intensive tasks in particular. We see MemLLM as an important step towards making LLMs more grounded and factual through memory augmentation.

The Associated Press reports on "a crewless robotic boat that had tried to retrace the 1620 sea voyage of the Mayflower" from the U.K. to Massachusetts' Plymouth Rock. And after five weeks it finally did reach North America. "The technology that makes up the autonomous system worked perfectly, flawlessly," an IBM computing executive involved in the project told the Associated Press. But "Mechanically, we did run into problems." It's especially disappointing because they'd tried the same voyage last year.

IBM's fully-autonomous edge AI-powered ship Mayflower has set off on its crewless voyage from Plymouth, UK to Plymouth, USA. The ship is named after the Mayflower vessel which transported pilgrim settlers from Plymouth, England to Plymouth, Massachusetts in 1620. On its 400th anniversary, it was decided that a Mayflower for the 21st century should be built. Mayflower 2.0 is a truly modern vessel packed with the latest technological advancements. Onboard edge AI computing enables the ship to carry out scientific research while navigating the harsh environment of the ocean--often without any connectivity.

Another ship called the Mayflower is set to make its way across the Atlantic Ocean this week, but it won't be carrying English pilgrims -- or any people -- at all. When the Mayflower Autonomous Ship leaves its home port in Plymouth, England to attempt the world's first fully autonomous transatlantic voyage, it will have a highly trained "captain" and a "navigator" versed in the rules of avoiding collisions at sea on board, both controlled by artificial intelligence (AI). The ship's AI captain was developed by Marine AI and is guided by an expert system based on IBM technologies, including automation software widely used by the financial sector. The technology could someday help crewed vessels navigate difficult situations and facilitate low-cost exploration of the oceans that cover 70 percent of the Earth's surface. Over its roughly three-week trip, the Mayflower sea drone will sail through the Isles of Scilly and over the site of the lost Titanic to land in Plymouth, Massachusetts, as the colonists on the first Mayflower did more than 400 years ago.

The world's first fully autonomous ship is set to make its maiden voyage across the Atlantic next month. Inspired by the ship that brought the Pilgrims to North America, 'Mayflower 400' will be guided by artificial intelligence rather than a human crew. If all goes well, it will depart from Plymouth, England on May 15 and arrive at Plymouth, Massachusetts, about 3,000 miles and two weeks later. The original Mayflower, which transported 102 Pilgrims and other passengers, took 10 weeks to reach its destination in 1620. Mayflower 400 was set to embark on its transatlantic cruise last September for the Mayflower's 400th anniversary, but was delayed because of the coronavirus pandemic.

The Mayflower is taking to the water in Plymouth harbor. It's not the ship that left this southwest England port 400 years ago carrying Pilgrim settlers to America. The sleek vessel being readied Tuesday for its official launch has no passengers, no crew -- but like its predecessor, an ambitious mission. The 50-foot (15-meter) trimaran has "no one on board, no captain, no place to eat, no place to sleep," said Brett Phaneuf, co-director of the Mayflower Autonomous Ship project. The ship is set to follow in its forebear's footsteps by crossing the Atlantic from Plymouth, England, to Plymouth, Massachusetts, this time on a marine research trip.

IBM has joined efforts to build a fully autonomous ship capable of journeying across the Atlantic. The tech giant announced on Wednesday that it will participate in a global consortium designed to help materialize the ship, called the Mayflower, which will make the same voyage as its namesake that originally carried pilgrims to what's now the continental US. Once completed, the self-navigating craft is slated to make a 3,000-mile journey across the Atlantic on September 2020 - a date that coincides with a 400-year anniversary for the original Mayflower's voyage. Using artificial intelligence designed by IBM, the Mayflower is being built to adeptly avoid ocean obstacles and complete it's journey from Plymouth England to Plymouth Massachusetts free of any human intervention. IBM and partner Promare, a marine research organization, are positioning the ship as a way of advancing research via reducing barriers like cost and human resources.

PLYMOUTH, Mass., September 9, 2019 -- Microway, a leading provider of computational clusters, servers, and workstations for AI and HPC applications, announces it has provided Oregon State University with six NVIDIA DGX-2 supercomputer systems, deployment services, and bringup expertise. Each DGX-2 packs 16 fully connected Tesla V100 GPUs, giving Oregon State a linked network of the world's most powerful AI systems powered by 96 GPU accelerators. The new, massively increased computing capabilities at the College of Engineering resolved a significant campus hardware gap and helped support cutting-edge research on medical imaging, nuclear science, bridge construction, robotics, and driverless vehicles. When planning expanded capability, university faculty and administrators determined they needed enough GPU capacity to serve the diverse needs of undergraduate classes and research workloads, plus lightning-fast storage. The University selected the NVIDIA DGX-2 platform for its immense power, technical support services, and the Docker images with NVIDIA's NGC containerized software.

Until last year, Brandon Tory, a senior software Artificial Intelligence engineer at Google and rapper, led a secret double life. Raised in Brockton, Massachusetts, a neighborhood known for crime and drugs, he lived with his family in a shelter as a teenager. Tory knew he wanted to be some kind of scientist, but also had a passion for music. He wanted to have a huge impact in both creativity and science. At University of Massachusetts Amherst he studied computer engineering and then worked as a senior Apple engineer in Cupertino.